in shape and of the size of a chiang bean (dolichos sinensis). Ch'ien means «cash»,
«copper coin», and is used by both I-ching and Li-yen to render Skr. paṇa (BAGCHI,
Deux lexiques sanskrit-chinois, 50, 229), but the ch'ien here intended must have been
much below the paṇa in value. This is proved by another gloss of Hui-lin, ch. 26
(磁, x, 2b) : «Paṇa means a copper coin (t'ung-ch'ien); 16 paṇa make one kārṣāpaṇa.» So
the modern scale of 16 paṇ (<paṇa) to one kāhan (<kārṣāpaṇa) goes back at least to the
beginning of the ninth century. On the other hand the traditional equivalence of the kārṣāpaṇa
to one karṣa of 16 māṣa is not in agreement with the scale adopted in the Mahāvyutpatti where
the kārṣāpaṇa is said to be worth 1600 cowries, and consequently to represent not 16 but
20 māṣaka (= māṣa) of 80 cowries each. This value of 1600 cowries to the kārṣāpaṇa is expressly
stated by I-ching, in a note added to ch. 6 of his translation of the Suvarṇaprabhāsa, to have
been the one universally used in Magadha in his time, i. e. in the second half of the seventh
century (cf. 磁, ix, 23b). The two values of 1280 and 1600 cowries for the kārṣāpaṇa have
been known to Chinese commentators, and they are also given by Tibetan sources (cf. Sarat
CHANDRA DAS, Tibetan-English Dict., 300). Such variations in value may be partly due
to fluctuations in the modes of reckoning with 4 or with 5 or, still more, to attempts
to pass from the vigesimal reckoning expressed by the unit of 80 cowries to a purely
decimal system based on a unit of 100. Other translators, probably misinformed, and because
of the ambiguity of ch'ien (which is either 0.1 of an ounce, or a coin in general, or a Chinese
copper cash) have understood the paṇa as being one copper cash (ch'ien) and the kārṣāpaṇa as
being 16 ch'ien, or have even merely rendered kārṣāpaṇa as ch'ien. It must be owing to some
misunderstanding of this kind that grar-ša-pa-ni, i. e. kārṣāpaṇa, is given as a synonym of iba'u,
«cowry», in KOWALENSKI, 281. I-ching further remarks that, in the text which he translates and
which speaks of «100 kārṣāpaṇas», one ms. gives «100 陳 那 磁 ch'ên-na-lo», i. e. «100 dīnāra»
or «gold coins»; a substitution rendered possible by the fact the kārṣāpaṇa, being in principle a
certain weight, could, theoretically at least, be used of gold as well as of silver. More details
are available on the kārṣāpaṇa; but I cannot discuss them here, and must refer the reader to
I-ching (loc. cit.), Hsüan-ying, ch. 22 (磁, vii, 86a), Hui-lin, ch. 13 (ib., viii, 97a), ch. 29 (ib.,
viii, 180a), ch. 71 (ib., x, 2b), K'o-hung, ch. 2 (ib., i, 21b), Bongo jiten, 114-115 and 454-455,
and ODA TOKUNO, 210.
The scale of the cowries in Yün-nan was 4, 16 (4 × 4), and 80 (16 × 5), but in modern
India it was 4 and 80 (4 × 20). The «four» basis of the reckoning also occurs in other ways in
the values attributed to the kārṣāpaṇa : for since it was worth 400 ch'ien or 1600 cowries, the
«ch'ien» of the first text of Hui-lin must have been worth four cowries, and must render not of
course paṇa as usual, but gaṇḍaka > ganda, i. e. four cowries. At the same time, the number
«400» itself, i. e. 20 × 20, supports PRZYLUSKI's view of a vigesimal reckoning, which also appears
in the higher values : 400 karṣa make one tulā (= Hind. tolā; cf. Hobson-Jobson², 928; FERRAND,
in JA, 1920, ii, 293 [where I doubt that we should start form Pers. tōlah], 294). I do not know
the authority for GILES's statement (Glossary of Reference, 63) that 200 cowries make «one āna
or about three halfpence»; the āna was usually worth 4 paṇ, i. e. 320 cowries.
Both paṇa and karṣa have become usual terms in foreign trade.
IIIF Color HighRes
